Trait Anger: STAXI
Students (M = 9.23, SD = 2.31) scored higher than employees (M = 8.24, SD = 2.32) on the Trait Anger Reaction subscale, F(1,281) = 11.53, p < .001, with omega2 = .028 (a small to moderate effect). This effect was maintained in a subsequent ANCOVA with age entered as a covariate, F(1,281) = 4.28, p < .03. This finding indicates that students have a greater tendency to respond with anger to an evaluative situation ("I feel infuriated when I do a good job and get a poor evaluation") than do employees. There were no significant differences between students and employees on the Trait Anger Temperament subscale. Students did not rate themselves as "quick tempered" more than did employees. In addition, there were no differences by age on the Trait Anger Temperament subscale.
Anger Labeling, Frequency, and Duration
The hypothetical situation appeared to elicit a definite emotional response in the participants. The majority of each group indicated that they would "likely" or "very likely" experience anger or frustration (students, 87%; employees, 65%). The student group responded with greater emotional intensity than did the employee group, chi2(1,N = 281) = 18.92, p < .001. Across the two groups, more participants selected "frustrated" (students, 67%; employees, 77%) than "angry" (students, 33%; employees, 23%) to describe their emotional reaction to the hypothetical situation, chi2(1,N = 281) = 5.28, p < .02.
We compared the students and employees on their likelihood of experiencing anger and expressing anger in response to the hypothetical situation. Students indicated greater likelihood of experiencing anger than did employees, chi2(1,N = 281) = 13.7, p < .001. However, when age was entered with work status in a logistic regression, the work status effect was not maintained. Students also indicated greater likelihood of expressing their anger than did employees, chi2(1,N = 281) = 28.46, p < .001. In a subsequent logistic regression analysis, work status continued to make an independent contribution to anger expression (beta = .42, p < .001) even when age was entered. Thus, work status was a better predictor than developmental period for anger expression but not for the experience of anger.
The responses to the hypothetical situation provided a sharp contrast between typical and situation-specific anger duration. When participants were asked to describe their typical response to an anger-provoking situation, the most frequent anger duration ranged from "a few minutes" to "up to a couple of hours" (students, 83%; employees, 74%). However, in response to the specific work/school-related situation, 34% of the students indicated an anger duration of "several days" and 58% of the employees indicated an anger duration of either "a week or more" or "several days."
In their descriptions of emotional reactions to the hypothetical situation, participants often reported more than one emotion. Table I presents the emotions identified as coexisting with anger. The emotions stereotypically associated with anger (i.e., frustrated and annoyed) are presented in the upper panel of the table. The emotions less stereotypically associated with anger (i.e., sad and hurt) are presented in the lower panel. Thirty-nine percent of the employees and 26% of the students reported feeling sad and mad at the same time. There were no significant differences between employees and students on frequency of a particular coexisting emotion.
Contextual Influences on Anger Experience and Expression
We examined situational characteristics that moderate an individual's likelihood of experiencing and/or expressing an emotion. A moderator may be the nature of a relationship (e.g., friend, family member) or the context of a potential anger-provoking situation (e.g., public place, one's overall stress level). For example, a woman's likelihood of becoming angry about not receiving a raise may be impacted by trust in the supervisor to rectify the situation in the future. In this case, trust in the supervisor moderates the woman's anger experience.
When faced with a situation in which a goal was blocked (hypothetical situation), participants first indicated the extent to which various factors would increase, decrease, or have no effect on feeling angry. Next, they indicated which factors would increase, decrease, or have no impact on expressing anger. The moderators of anger experience were similar for students and employees. Both groups were more likely to experience anger if the situation had happened before, presumably with an unsatisfactory outcome. Conversely, if reasons for denial appeared legitimate or if there was a promise of future reward, both groups were less likely to feel angry about the blocked goal. Very few students and employees (less than 10%) indicated that the various factors would have no effect on their feelings or actions.
Yet, there were differences between the two groups on two factors. Employees were more likely to experience anger than were students "if others received the benefit" (e.g., had received the raise or the scholarship), chi2(1,N = 281) = 15.2, p < .001. On the other hand, trust in one's supervisor/administrator was less likely to reduce students' anger compared with employees' anger, chi2( 1,N = 281) = 12.2, p < .001. In the student scenario, the context is trust in a school administrator rather than a work supervisor. Hence, the difference on this item may be related to the low likelihood that the student knows the administrator well enough to have developed a trust relationship. In this case, the issue of personal trust may be a less salient factor for the student population. Age was not a factor in either comparison.
In terms of the moderators of anger expression, a shared pattern of influences was found. The majority of students and employees reported that "the importance of the situation" (students, 88%; employees, 95%) as well as "high personal stress" (91% and 95%, respectively) would increase the likelihood of expressing anger. Both groups were less likely to express anger if "the anger was not justified" (students, 83%; employees, 80%), "the situation occurs in public" (86% and 90%, respectively), or "the target person will feel hurt" (86% and 82%, respectively). Gender of the target did not appear to have an effect on the likelihood of anger expression for either group.
In sum, there were similarities in trait anger characteristics among the participants. Some employees and students reported that feelings of sadness and hurt coexist with anger. Employees and students were equally likely to respond to the blocked goal with anger/frustration, and common factors moderated the experience of anger. As for the differences, students reported more intense reactions to an anger-provoking situation than did employees. Students also indicated greater likelihood of expressing anger, even when age was controlled.
The present study examined the anger and expression of females who differed by work status and age. For both students and employees, issues of violated trust and unfair treatment increased anger. But even when angered, concern about the target's feelings or concern about being observed by others reduced the likelihood of expressing that anger. Further, anger often coexisted with "more acceptable feminine emotions," such as sadness. These commonalities point to the effectiveness of the socialization of females regarding emotional expression. Yet, there were differences between the students and employees. Independent of age, students were more likely to express anger in response to the hypothetical situation, and responded more intensely, compared with employees. These results offer some support for the hypothesis that a "less feminine" work environment (e.g., academia) may be conducive to anger expression.
Consistent with previous research findings, violated expectations for fair treatment increased the likelihood that the females would respond with anger. However, the anger could be moderated (e.g., if promises have been made to restore equity in the future, or if there is trust in the supervisor/administrator). Consistent with the theme of equity expectations, women would become angry if the situation had occurred before, presumably with lack of a satisfactory resolution, or if others with similar qualifications were rewarded. Thus, a common pattern emerges in which a female experiences anger not solely as a reaction to a blocked goal, but after careful consideration of the relative fairness of the situation. This pattern is congruent with socialization for the feminine role. Girls and women are socialized to maintain harmonious relationships and are likely to expect others to do the same. Therefore, when a person violates a relationship by not providing fair treatment, a woman evaluates her own anger as justifiable.
However, anger experience does not necessarily translate into anger expression. The childhood socialization practices that emphasize harmonious relationships again play a role. In general, females are concerned about the potential negative impact of anger expression on others and thus will temper their anger expression to avoid hurting the feelings of the target person. Further, beginning in early childhood, females understand that showing their anger will likely result in social censure. This is evident in the findings presented here: participants indicated that they are less likely to express their anger if the situation occurs in public. The potentially negative evaluation by others influences the decision to express anger.
Differences by Work Status
With respect to stable contexts, work status plays a role in anger expression. Although all participants were equally likely to experience anger, those in an office work environment were less likely to express their anger overtly. Moreover, the students were more likely to react with anger to an evaluative situation. This characteristic may be related to the academic role, in which constant evaluation may make students more sensitive to negative feedback. This difference in anger expression is also consistent with the students' greater willingness to express anger when faced with a blocked goal.
We can extrapolate from Stein's work on the function of an emotional response to a blocked goal. If a woman assumes that a goal (e.g., receiving a raise or a scholarship) is nonattainable, her emotional response is likely to be sadness or resignation, but not anger. However, if the goal is perceived as attainable, the predicted emotional response would be anger. Students may perceive a scholarship as attainable, perhaps even an entitlement, and therefore respond with anger if the scholarship is not received. Further, employees may view their work competence as independent of an increase in salary, whereas students may view a scholarship as tangible evidence of their academic success. Thus, it is possible that some of the differences in the responses to the hypothetical situation may be related to female students' and employees' investment in the work role and how they define success in that role.
Certain limitations should be noted. First, based on the responses to the hypothetical anger-provoking situation, it is possible that the deferred raise was not as highly emotive an issue for the employees as the deferred scholarship was for the students. Second, the two groups differed in age as well as work status and thus we are restricted in our ability to consistently ascribe group differences solely to work or developmental status. Finally, it is possible that by describing the project as research on women's anger, we may have inhibited women from reporting additional emotions. Using a more neutral label, such as "upset," might have encouraged the description of a greater range of emotions.
Nevertheless, we found that similar situations provoke anger in females across work status and age. In addition, the possibility of negative social evaluation constrains anger expression in young students and older office employees alike. However, females in more gender-stereotypic work environments (i.e., office employees) are less likely to express their anger when compared to females in an environment with gender-balanced expectations (i.e., university students).
Some females may label their emotional experiences, as well as behave, in ways that avoid the appearance of anger. For example, sadness or hurt feelings may reflect an appreciation of the social costs of anger expression, or may reflect the perceived failure to successfully negotiate conflict. Whether discomfort with the label of anger is a pervasive phenomenon merits further scrutiny, because there are implications for effective problem-solving in a work environment. If women do not acknowledge their anger or have little practice in effectively expressing their anger, work productivity can suffer.
The lengthy anger duration in response to the hypothetical situation suggests a ruminative cognitive style. Women who ruminate play out the situation over and over in their minds. Such a pattern does not resolve the anger-provoking situation or the uncomfortable emotions. We also need more in-depth data from women who rarely become angry or who indicate a lack of anger response. One focus would be to explore the possibility of a connection with indirect anger/aggression. Women who have devised a plan for covert retaliation (indirect aggression) may not see themselves as experiencing anger. In conclusion, we need to be cognizant of the interactive influences of socialization and work role expectations on women's ability to describe and manage their anger.
- Hatch, Holly & Deborah K. Forgays, A Comparison of older adolescent and adult females’ responses to anger-provoking situations, Adolescence, Fall 2001, Vol. 36, Issue 143.
Reflection Exercise #2
The preceding section contained information
about the results and conclusion of a study on older adolescent and adult females’ responses to anger-provoking situations. Write
three case study examples regarding how you might use the content of this section
in your practice.
According to Hatch & Forgays, what three factors made both groups of students and employees less likely to express anger? Record the letter of the correct answer